Distribution: normal, binomial and Poisson from one screen
Skip the data table entirely: name the curve, type the x the question asks about, and read either the height or the area up to it.
8 min read Intermediate updated 2026-11-07
Distribution is the Statistics family with the data taken out. You supply the shape instead: which model, its centre and spread, and the one value the question asks about. The machine then returns either the height of the curve at that x or the area under it up to that x. Nothing is stored between runs, so there is no stale column to clean out first.
Open a distribution and enter the numbers
Keystrokes for a Normal curve with μ = 60 and σ = 5
- 1 MENU
Open the application menu.
- 2 ▼ ▶
Move the highlight onto the Distribution icon.
- 3 =
Open the application.
- 4 ▼ =
Choose the model. Follow the label rather than a number: the list is Normal, Binomial, Poisson and the rest, and its order differs between EX models.
- 5 ▼ =
Choose which of the two entries you want - the value at x, or the cumulative area up to x. Every model in this application offers both.
- 6 6 5 =
The x field: 65, the mark the question asks about.
- 7 5 =
The spread field: σ = 5.
- 8 6 0 =
The centre field: μ = 60.
- 9 =
Confirm and read the result, then press AC to leave. The screen repeats the field names back, so check them once before you write anything down.
Normal: the area, not the height
Take marks that follow a Normal curve with μ = 60 and σ = 5. Asking for the area up to 65 is the same question as 'what share of candidates scored 65 or less', and 65 is exactly one σ above the mean.
| Question | Screen | How to check it |
|---|---|---|
| P(X ≤ 65) | 0.841345 | one σ above the mean: about 84% of the area |
| P(X ≤ 55) | 0.158655 | the mirror image, 1 − the row above |
| P(55 ≤ X ≤ 65) | 0.68269 | the middle strip - the 68% of 68-95-99.7 |
| height at x = 65 | 0.048394 | a density: no width, so no probability |
| height at x = 60 | 0.079788 | the peak, because the curve is highest at μ |
- P(X > 65) = 1 − 0.841345 = 0.158655. A Normal curve is symmetric about μ, so the right tail above 65 and the left tail below 55 are the same number - a two-second way to catch a σ and μ typed the wrong way round.
- Subtract before you round: the middle area is the difference of the two full-digit areas the machine is holding, not of the six-decimal numbers printed above.
- The figures here are rounded to six decimals. Your screen carries more digits, and how many you see is a display setting rather than a calculation.
Binomial and Poisson: counts, not measurements
Both of these work on whole numbers of successes, and both need the height-versus-area choice settled before you type. Ten flips of a fair coin is the smallest case where the whole answer can be checked on paper.
| Model | Question | Screen |
|---|---|---|
| Binomial n = 10, p = 0.5 | exactly 3 heads | 0.1171875 |
| Binomial n = 10, p = 0.5 | 3 heads or fewer | 0.171875 |
| Poisson λ = 3 | exactly 2 arrivals | 0.2240418 |
| Poisson λ = 3 | 2 arrivals or fewer | 0.4231901 |
The binomial pair is arithmetic: the coefficients of (a + b)¹⁰ run 1, 10, 45, 120, 210 and so on, so P(X = 3) = 120 ÷ 1024 = 0.1171875 and P(X ≤ 3) = (1 + 10 + 45 + 120) ÷ 1024 = 176 ÷ 1024 = 0.171875.
Input
176 ÷ 1024
Result
0.171875
The denominator is 2¹⁰, one per outcome. If the screen is showing a fraction, S⇔D turns the same result into this decimal.
- The means come free: a binomial is np = 5 heads, a Poisson is λ = 3 arrivals. An answer a long way from those numbers deserves a second look.
- The spread is the row after the mean - √(np(1 − p)) = 1.581139 for the coin, √λ = 1.732051 for the arrivals - and it is the same kind of number the Normal curve asked for as σ.
- Binomial counts successes out of a fixed n; Poisson counts events in a stretch of time or space with no n to point at. Picking the wrong model is the one mistake the machine cannot see.
- With λ = 3 exactly, P(X = 2) and P(X = 3) are the same 0.2240418: a Poisson curve whose mean is a whole number has two tallest bars, so 'the most likely count' is a question with two answers.
Was this guide helpful?
Tell us what was unclear, missing or wrong. A concrete sentence is more useful than a rating.